dsh-linghun
Adds a cognitive loop, hippocampus three-layer memory, and closer decision discipline to DeepSeek Harness agents.
Install
$ dsh plugin --profile web add dsh-linghunPlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Three-layer hippocampus memory featuring warm buffers, episodic archives, and cold summary injection
- Chronological ledger (journal_read) tracking timestamped conversation history with second-level range lookups
- Automated end-of-turn memory assessment and threshold-based auto-consolidation without manual prompts
- Self-evolving soul cards and persona management stored in plain Markdown files
- Confidence-weighted memory indexing and degradation for overturned entries
- Enforces a three-rule judgment discipline preventing model hallucination and persona fabrication
Useful For
- Eliminating endless agent planning by enforcing prompt-level closer architecture and concrete execution criteria
- Maintaining long-term conversational memory across sessions via structured Markdown archives
- Building self-updating agent personas that retain preferences and stable traits over time
Who It Fits
- DeepSeek Harness developers seeking persistent memory and cognitive loop frameworks
- Agent builders requiring rigorous verification and closing disciplines for autonomous agents
Documented Limitations
- Requires DeepSeek Harness >= 0.1.0-rc.7 (< 0.3.0) and Node.js >= 20.18
- Designed specifically for the DSH 'web' profile
- Relies on local file system read/write access under $DSH_HOME/linghun/
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 1 runtime, 0 development, 3 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as NOASSERTION.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- syyr1987/dsh-linghun
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 377e130 · 2026-10-16
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 3GitHub stars
- Forks
- 0
- Open issues
- 1
- Last commit
- 2026-10-10
- Last release
- 2026-10-10
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Native SQLite-backed task management and project board bundle with Web UI and Agent workflow tools for DeepSeek Harness.
View plugin →Persistent multi-model agent teams with dynamic DAG orchestration, run monitoring, and token insights for DeepSeek Harness.
View plugin →Second-model AI auto-review for DeepSeek Harness approval requests with fail-closed security.
View plugin →Import and resume conversation histories from 25+ AI coding agents directly into DeepSeek Harness.
View plugin →